Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments
Recently, there has been significant interest in applying machine-learning (ML) techniques to the automated analysis of X-ray scattering experiments, due to the increasing speed and size at which datasets are generated. ML-based analysis presents an important opportunity to establish a closed-loop f...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
International Union of Crystallography
2023-11-01
|
| Series: | Journal of Synchrotron Radiation |
| Subjects: | |
| Online Access: | http://scripts.iucr.org/cgi-bin/paper?S160057752300749X |
| _version_ | 1797361746199969792 |
|---|---|
| author | Linus Pithan Vladimir Starostin David Mareček Lukas Petersdorf Constantin Völter Valentin Munteanu Maciej Jankowski Oleg Konovalov Alexander Gerlach Alexander Hinderhofer Bridget Murphy Stefan Kowarik Frank Schreiber |
| author_facet | Linus Pithan Vladimir Starostin David Mareček Lukas Petersdorf Constantin Völter Valentin Munteanu Maciej Jankowski Oleg Konovalov Alexander Gerlach Alexander Hinderhofer Bridget Murphy Stefan Kowarik Frank Schreiber |
| author_sort | Linus Pithan |
| collection | DOAJ |
| description | Recently, there has been significant interest in applying machine-learning (ML) techniques to the automated analysis of X-ray scattering experiments, due to the increasing speed and size at which datasets are generated. ML-based analysis presents an important opportunity to establish a closed-loop feedback system, enabling monitoring and real-time decision-making based on online data analysis. In this study, the incorporation of a combined one-dimensional convolutional neural network (CNN) and multilayer perceptron that is trained to extract physical thin-film parameters (thickness, density, roughness) and capable of taking into account prior knowledge is described. ML-based online analysis results are processed in a closed-loop workflow for X-ray reflectometry (XRR), using the growth of organic thin films as an example. Our focus lies on the beamline integration of ML-based online data analysis and closed-loop feedback. Our data demonstrate the accuracy and robustness of ML methods for analyzing XRR curves and Bragg reflections and its autonomous control over a vacuum deposition setup. |
| first_indexed | 2024-03-08T15:58:50Z |
| format | Article |
| id | doaj.art-bc42c255ce3747ecbedc3e9071da5fd3 |
| institution | Directory Open Access Journal |
| issn | 1600-5775 |
| language | English |
| last_indexed | 2024-03-08T15:58:50Z |
| publishDate | 2023-11-01 |
| publisher | International Union of Crystallography |
| record_format | Article |
| series | Journal of Synchrotron Radiation |
| spelling | doaj.art-bc42c255ce3747ecbedc3e9071da5fd32024-01-08T14:37:41ZengInternational Union of CrystallographyJournal of Synchrotron Radiation1600-57752023-11-013061064107510.1107/S160057752300749Xju5054Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environmentsLinus Pithan0Vladimir Starostin1David Mareček2Lukas Petersdorf3Constantin Völter4Valentin Munteanu5Maciej Jankowski6Oleg Konovalov7Alexander Gerlach8Alexander Hinderhofer9Bridget Murphy10Stefan Kowarik11Frank Schreiber12Institut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, GermanyInstitut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, GermanyPhysikalische und Theoretische Chemie, Universität Graz, Heinrichstrasse 28, 8010 Graz, AustriaInstitut für Experimentelle und Angewandte Physik, Universität Kiel, Leibnizstrasse 19, 24118 Kiel, GermanyInstitut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, GermanyInstitut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, GermanyESRF – The European Synchrotron, 71 Avenue des Martyrs, CS 40220, 38043 Grenoble Cedex 9, FranceESRF – The European Synchrotron, 71 Avenue des Martyrs, CS 40220, 38043 Grenoble Cedex 9, FranceInstitut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, GermanyInstitut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, GermanyInstitut für Experimentelle und Angewandte Physik, Universität Kiel, Leibnizstrasse 19, 24118 Kiel, GermanyPhysikalische und Theoretische Chemie, Universität Graz, Heinrichstrasse 28, 8010 Graz, AustriaInstitut für Angewandte Physik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, GermanyRecently, there has been significant interest in applying machine-learning (ML) techniques to the automated analysis of X-ray scattering experiments, due to the increasing speed and size at which datasets are generated. ML-based analysis presents an important opportunity to establish a closed-loop feedback system, enabling monitoring and real-time decision-making based on online data analysis. In this study, the incorporation of a combined one-dimensional convolutional neural network (CNN) and multilayer perceptron that is trained to extract physical thin-film parameters (thickness, density, roughness) and capable of taking into account prior knowledge is described. ML-based online analysis results are processed in a closed-loop workflow for X-ray reflectometry (XRR), using the growth of organic thin films as an example. Our focus lies on the beamline integration of ML-based online data analysis and closed-loop feedback. Our data demonstrate the accuracy and robustness of ML methods for analyzing XRR curves and Bragg reflections and its autonomous control over a vacuum deposition setup.http://scripts.iucr.org/cgi-bin/paper?S160057752300749Xmachine learningreflectometryautonomous experimentsbeamline controlxrrclosed-loop control |
| spellingShingle | Linus Pithan Vladimir Starostin David Mareček Lukas Petersdorf Constantin Völter Valentin Munteanu Maciej Jankowski Oleg Konovalov Alexander Gerlach Alexander Hinderhofer Bridget Murphy Stefan Kowarik Frank Schreiber Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments Journal of Synchrotron Radiation machine learning reflectometry autonomous experiments beamline control xrr closed-loop control |
| title | Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments |
| title_full | Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments |
| title_fullStr | Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments |
| title_full_unstemmed | Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments |
| title_short | Closing the loop: autonomous experiments enabled by machine-learning-based online data analysis in synchrotron beamline environments |
| title_sort | closing the loop autonomous experiments enabled by machine learning based online data analysis in synchrotron beamline environments |
| topic | machine learning reflectometry autonomous experiments beamline control xrr closed-loop control |
| url | http://scripts.iucr.org/cgi-bin/paper?S160057752300749X |
| work_keys_str_mv | AT linuspithan closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT vladimirstarostin closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT davidmarecek closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT lukaspetersdorf closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT constantinvolter closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT valentinmunteanu closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT maciejjankowski closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT olegkonovalov closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT alexandergerlach closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT alexanderhinderhofer closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT bridgetmurphy closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT stefankowarik closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments AT frankschreiber closingtheloopautonomousexperimentsenabledbymachinelearningbasedonlinedataanalysisinsynchrotronbeamlineenvironments |